|Assoc Prof Cham Tat Jen |
Division of Visual & Interactive Computing
School of Computer Engineering
College of Engineering
Phone: (+65)6790 6230
- PhD University of Cambridge 1996
- BA(Eng)(Hons) University of Cambridge 1993
|Tat-Jen is an Associate Professor in the School of Computer Engineering, and the Director of the Centre for Multimedia and Network Technology (CeMNet). He obtained his BA in 1993 and PhD in 1996 from the University of Cambridge. He was a Jesus College Research Fellow at Cambridge 1996-97, before joining DEC/Compaq Research Lab in Boston as a research scientist 1998-2001. While at NTU, he was concurrently a Faculty Fellow in the Singapore-MIT Alliance Computer Science Program 2003-06. Tat Jen has made key contributions in computer vision and projector-camera systems, receiving overall best paper prizes at PROCAMS'05, BMVC'94, and in particular at ECCV'96 (now named the Longuet-Higgins prize). Tat Jen is an inventor on 8 patents and filings, and his papers have been cited over 800 times. He is on the editorial board for IJCV and an Area Chair for ICCV'05, the top journal and conference in his field respectively. Tat Jen has been interviewed on national media in relation to SMA projects that he jointly supervised.
|Cham Tat Jen has worked on a number of research areas within computer vision.
(1) Matching and registration in images. 1994-96: matching partially observed image curves under different perspective projections. 1998: wide-baseline image mosaicing. 2002-present: matching between approximate 3D models and visible/thermal imagery. 2005-present: matching images related by high-level abstract concepts. Achievements: best paper prize in BMVC'94, best paper prize in ECCV'96; CVPR'97 oral, CVPR'98, T-PAMI (1999), CVPR'05, CVPR'07, CVPR'08; 2 DSTA grants worth $630K.
(2) Human tracking, modeling and understanding. 1998-2001: detailed tracking of humans as articulated figures in monocular video. 2003-: human activity recognition. Achievements: CVPR'99 oral paper cited 250+ times, 2 ICCV'99 orals with one cited 100+ times, CVPR'01 oral paper cited 50+ times; 6 US patents; invited talk at Pixar in 1999; various aspects of the framework has been implemented elsewhere, including at MIT, INRIA and IBM Research.
(3) Projector-camera systems. 2001-: novel interfaces using projector-camera systems, including an active shadow-elimination system for projector displays, and interactive interior relighting system simulating multiple light sources with a single projecor. 2003-2005: photometric calibration. 2005-2006: oblique out-of-focus compensation. Achievements: PROCAMS'05 best paper prize; CVPR'01, CVPR'03, CVPR'06, T-VCG (2007); 1 US patent and 1 ITTO-sponsored filing; material used in MDA IDM video (on MDA IDM website) and TeIN2 video; consultation with MINDEF Future Systems Directorate.
(4) Accelerated online machine learning. 2004-: online accelerating speed of inference during the processing of online learning for vision-based object detectors. 2007-: fast detection and learning for Adaboost-based face detectors. Achievements: CVPR'07 oral, ICCV'07 oral (note: 1 of only 2 1st-2nd author pairings in the world to achieve 2 oral papers in CVPR and ICCV 2007, out of 2500+ submissions), CVPR'08; PhD student received ICCV'07 student travel grant; PREMIA 2nd best student paper prize in 2007 (prized awarded for top Singapore student papers published in any pattern recognition-related journal or conference); open source PyCV package released and has been downloaded worldwide many times.
|Research Grant |
- Defence Science and Technology Agency (2002-2014)
- MDA-IDMPO (2012-)
- NRF CREATE (2011-) [by National Research Foundation (NRF)]
|Current Projects |
- Advanced Technologies for 3D Capture, Communication and Display
- Human Gait Recognition with Statistical Learning
- Novel Scene Acquisition and Display Technologies for a Next-Generation Room-Based Telepresence System
- Project Tacrea
- Project image understanding - developing automatic target detection and recognition capability
- SMART Space
- Singapore – ETH Centre (SEC) for Global Environmental Sustainability: Future Cities Laboratory (FCL) programme
- Tat-Jen Cham, Ciptadi Arridhana, Wei-Chian Tan, Minh-Tri Pham, Liang-Tien Chia. (2010). Estimating Camera Pose from a Single Urban Ground-View Omnidirectional Image and a 2D Building Outline Map. IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
- Minh-Tri Pham Viet-Dung D. Hoang Tat-Jen Cham. (2008). Detection with Multi-exit Asymmetric Boosting. (pp. 1 - 8).
- D. Xu, T.J. Cham, S. Yan and S.-F. Chang. (2008). IEEE International Conference on Computer Vision and Pattern Recognition: Near Duplicate Image Identification with Spatially Aligned Pyramid Matching.
- Minh-Tri Pham Tat-Jen Cham. (2007, October). Fast Training and Selection of Haar features using Statistics in Boosting-based Face Detection. Paper presented at 11th IEEE International Conference on Computer Vision, Rio de Janeiro, Brazil.
- C.K.Yeo, L.T.Chia, T.J.Cham, R.Deepu. (2007). IEEE International Conference on Multimedia and Expo (ICME): Click4BuildingID@NTU: Click for Building Identification with GPS-enabled Camera Cell Phone.Beijing, China: IEEE.